smengcl opened a new pull request, #11702: URL: https://github.com/apache/ozone/pull/11702
Generated-by: Claude Code (Opus 5.5) ## What changes were proposed in this pull request? The `kubernetes (minikube)` check failed on master before any test ran. k3s started and accepted the resources, then its API server went away a few seconds later, and every later `kubectl` call failed with `open /etc/rancher/k3s/k3s.yaml: permission denied`. `install/k3s.sh` makes the kubeconfig readable with a one time `chmod` after installation, but k3s rewrites `k3s.yaml` with mode 0600 whenever the server starts. So after a k3s restart the check cannot reach the cluster again, even if k3s recovers. This PR passes `K3S_KUBECONFIG_MODE=644` to the k3s installer. The installer stores it in the k3s service environment, so k3s writes the kubeconfig readable on every start. The existing `chmod` is kept unchanged. ## What is the link to the Apache JIRA https://issues.apache.org/jira/browse/HDDS-17347 ## How was this patch tested? - Checked that the k3s installer persists `K3S_*` variables into `k3s.service.env`, and that k3s v1.35.1 reads `K3S_KUBECONFIG_MODE` for `--write-kubeconfig-mode`. - The failure needs a k3s restart, which cannot be triggered reliably in CI, so this is not a reproduction of the original failure. - Kubernetes checks in [CI on the fork](https://github.com/smengcl/hadoop-ozone/actions/runs/39839152372): in progress 🤖 Generated with [Claude Code](https://claude.com/claude-code) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
